Cardiovascular disease (CVD) affects approximately 77% of the elderly, community-dwelling Medicare population and is the largest driver of hospital admissions, deaths, and health expenditures in this group. Since 2006, the subsidized Part D drug benefit has improved the accessibility of needed medications for people in Medicare, but the specific effects on those with CVD are not well-known, in particular in terms of medication adherence, use of health services, and patient outcomes. Moreover, nothing is known about major trends in these outcomes among elderly Medicare enrollees with CVD in the eight years following Part D implementation. Using strong quasi-experimental methods and the nationally-representative Medicare Current Beneficiary Survey (2000-2013), we will identify the population of elderly Medicare patients with CVD during a 14-year period and evaluate immediate and longer-term changes in their utilization and clinical outcomes following Part D. Study measures will include demographic and clinical attributes, drug and non-drug utilization, cost-related medication nonadherence, acute hospital events, and perceived health status. For the period after Part D implementation, we will develop and examine new measures including patient cost-coping strategies, month-to-month adherence, and out-of-plan purchasing. For high-risk CVD patients (those with drug spending near or above the Part D donut hole coverage gap threshold), we will estimate changes in outcomes following the sudden 2011 improvement in brand name drug coverage within this coverage gap. In particular, we will examine changes in adherence and CVD treatment intensity for Part D patients in or near the coverage gap, by therapeutic drug class and by brand versus generic status.